MARINE ECOLOGY PROGRESS SERIES Vol. 96: 239-248.1993 Published June 29 Mar. Ecol. Prog. Ser. ~ Mass mortality of abalone Haliotis cracherodii on the California Channel Islands: tests of epidemiological hypotheses Kevin D. Lafferty, Armand M. Kuris Department of Biological Sciences and Marine Science Institute, University of California. Santa Barbara. Santa Barbara, California 93106, USA ABSTRACT. Along the coast of the California Channel Islands, USA, the abundant black abalone Haliotis cracherodii began to die in large numbers in 1985. This die-off has spread to most of the islands and still continues. Analysis of the changes in abalone abundance at numerous locations suggests that this is the result of an infectious disease spreading slowly from the south side of Santa Cruz Island. However, this might not be due to a directly transmitted disease and we have found no agent of infec- tlon As with many diseases, warm temperatures acted to increase mortality rates. Our results were not consistent with other hypotheses for the decline such as over-harvesting, pollution, or factors related to physiological stresses induced by the strong El Nirio of 1983-84, including warm temperatures, starva- tion and competition. INTRODUCTION Colodey 1983) caused by an amoeba (Jones & Scheibling 1985) and of S. franciscanus along the Mass mortalities of marine organisms are infrequent California coast in the 1970s (Pearse et al. 1977) also but important phenomena (Williams & Bunkley- caused by an infectious disease, the spectacular Williams 1990). When an abundant organism rapidly decline in Diadema antillarium populations in the and completely disappears from a region, the com- Caribbean region in the 1980s (Lessios 1988) and the munity structure and its organizational dynamics may near elimination of the starfish Heliaster kubiniji in the undergo dramatic changes. These mass mol-talities Sea of Cortez in the 1970s caused by infectious agents may start ecological processes leading to the type of (Dungan et al. 1982). For these events, population alternative stable states recognized by Sutherland declines were both rapid and widespread and the (1974). If the species undergoing such a crash had causative agent often remalned unknown. These cir- previously sustained a fishery, or was valuable for cumstances have prevented the careful analysis of the other considerations, the mass mortality assumes even geographic pattern of these mass mortalities, as biolo- greater importance for human affairs. gists generally became aware of them after the event Some examples of mass marine mortalities include: was nearly over and, usually, lacked background data the loss of eelgrass Zostera marina beds to wasting dis- to document the spread and extent of the population ease in the 1930s along the Atlantic shore of the United decline. States (DenHartog 1987), the devastation of vast areas Long-term changes in the abundance of black of oyster Crassostrea virginica beds due to Perkinsus abalone Haliotis cracherodii on the California Channel marinus infection in mid-Atlantic estuaries of the Islands (USA) reflect historical events. Evidence from United States (Perkins 1976), the crash of sea urchin Indian middens (Cox 1962, Muratto 1984) shows that Strongylocentr-otus droebachiensis populations on the Native Am.ericans heavily fished the abundant black eastern coast of North America in the 1980s (Miller & abalone populations. In the early 1800s, the removal of Mar. Ecol. Prog. Ser. 96- 239-248, 1993 the human population (Johnson 1982, Douros 1985) Diablo Cove (Steinbeck et al. 1992) but not at other and the extirpation of sea otter populations by Russian mainland locations where long-term observations have fur hunters (VanBlancom & Estes 1988) virtually elimi- been available (Point Loma and the Palos Verdes nated vertebrate predatory pressure on black abalone Peninsula). populations. Other abalone predators, such as the Shepherd & Breen (1992) discuss a variety of factors California sheephead and the spiny lobster, have also responsible for past abalone mortalities in Cal~fornia declined (Tegner 1980, Tegner & Levin 1983). As a re- and elsewhere. Sometimes, it appears that human sult, throughout the twentieth century, black abalone factors were responsible. Coastal development, har- attained high population levels on the Channel Islands vesting and pollution have depleted coastal stocks in and became one of the most abundant large inverte- California (Haaker et al. 1986, Tegner 1989). In partic- brates in the middle and lower rocky intertidal zones ular, Young (1964) implicated sewer effluent at White's (Douros 1985). Indeed, it was typical to observe 3 or Point, Palos Verdes, for a shrunken foot condition 4 black abalone stacked on each other (Cox 1960, there. Copper discharged from the cooling system of Douros 1987). Since 1973, black abalone populations Diablo Canyon Nuclear Power Plant resulted in heavy in the Channel Islands have supported a substantial mortality to the adjacent population (Martin et al. commercial export fishery (Parker et al. 1992). 1977). Natural events also cause mortality. Storms can In 1985, commercial fishermen observed large result in abalone being crushed or torn off of the sub- numbers of empty shells and dying abalone on the strate (Cox 1962, Sainsbury 1982, Tegner & Butler southern shore of Santa Cruz Island. Similar die-offs 1989). In Australia, dinoflagellate blooms (R. Tarr in subsequently happened on Anacapa, Santa Rosa, Shepherd & Breen 1992) and the protozoan Perkinsus Santa Barbara, San Miguel, and San Clemente islands olseni are linked with high mortality (Lester et al. 1988, and, recently (summer of 1992), on San Nicolas Island O'Donoghue et al. 1991). In several locations, starva- (Fig. 1) (Haaker et al. 1992).There are no quantitative tion, due to the reduced availability of drift algae, has observations for Santa Catalina, nor for Islas also been documented (Ino 1966, Sakai 1962, Schmitt Coronados in Mexico. A similar pattern occurred at & Connell 1984, Tanaka et al. 1986). San Santa Barbara SanB Catalina .-SL -*San Nimlas San Clemenb Fig. 1 Channel Islands off the coast of southern California, showing the sites where abalone HaLiotis cracherodii populations were monitored. AP = Airport. CH = Cuyler Harbor, CP = Crook Pt., EE = Eel Pt.. EP = East Pt., EX = Exposed, FP = Ford Pt., FK - Fossil Reef, HP = Harris Pt., JL = Johnson's Lee, M = Middle. NE = North Eel Pt., NT = North West Talcott, OH = Otter Hat- bor, PR = Protected, RC = Rock Crusher, SL = Sea Lion, W = West. The solid part of each bar represents the remaining proportion (1991) of the initial population at each site Lafferty & Kuris: Mass mortality of abalone Although the pattern of declining fisheries landings efforts on changes in the density of sub-legal individu- suggests over-harvesting (Parker et al. 1992), there is als (<14.6 cm). This indicates fishery-independent an association between mortalities in the Channel mortality. It was also necessary to determine associa- Islands and a set of morphological conditions termed tions between the activities of the fishery and the die- withering syndrome (WS) for which the most striking off associated with M'S. To test the hypothesis that the feature is a withered foot. Only a quick and stealthy fishery is the cause of the mass mortality, we examined tug can remove a healthy abalone. However, weak- the relationship between fishing pressure and sub- ened abalone in late stages of the disease readily legal abalone density at many locations throughout the detach from the substrate. A weakened foot is not Channel Islands. If the fishery was indirectly responsi- diagnostic for WS because a withered foot is a common ble for the die-off, we predicted a negative association terminal symptom of several abalone diseases. Other between fishing intensity and changes in sub-legal aspects of WS include color changes of the foot and abalone density. Further, if WS is an entity spread by epipodium and reduced gonadal size (Haaker et al. human activities we expected that heavily fished 1992, Kuris et al. unpubl.). Sites suffering from WS areas, or those close to major ports, would become have abnormally high numbers of fresh, empty shells affected at early dates. from all sizes of recently dead individuals (Haaker et Some evidence suggests that El Nifio periods with al. 1992).These observations indicate that declines in warm water temperatures were the initial cause of WS abalone densities are, at least partially, independent of and the resultant mortality (Tissot 1988, 1990, Davis et the commercial or sport fishery, and may be attribut- al. 1992).A variant of this hypothesis is that because El able to WS (because landed abalone must be of legal Nirio events reduce kelp abundance (Tegner & Dayton size and have an intact shell). 1987), and drift kelp is the main food source of black Although the cause of WS has been elusive, there abalone (Cox 1962, Leighton & Boolootian 1963, are a number of proposed hypotheses. While the most Douros 1987), black abalone may have starved to complicated hypothesis involves 16 interacting causal death (Tissot 1988. 1990). Evidence in support of the factors (Davis et al. 1992), the only explicitly tested El Nino hypothesis includes: (1) the die-offs began hypothesis is that a recently discovered, as yet unde- soon after the strong El Nino of 1983-84, (2) in 1988, scnbed, coccidian protozoan associated with kidney mortality occurred in association with warm water damage causes WS (Steinbeck et al. 1992). Rejection of discharged from Diablo Canyon Nuclear Power Plant this infectious agent as the cause of WS stems from the (Steinbeck et al. 1992), and (3) abalone at Ario Nuevo, observation that it is present in all of the black abalone presumably north of the influence of El Nirio, did not sampled (it may be present at higher intensities in suffer from WS (Tissot 1988). However, the occurrence healthy animals than in sick animals), and there is no of WS after El Nino could have been happenstance and association between coccidian intensity and increased warm water may not have been the sole causal factor WS pathology (Steinbeck et al.